Two-dimensional Objects In The Active Shape Model (asm) Medical Image-based Positioning Methods Discussed

Localization of an object in an image is always one of the most typical research topics in the fields of image processing and computer vision. The technology of object localization is a new application in the field of digital image processing and has been taken widely use of in the areas of military,medical and computer sciences,such as precise targeting weapons,computer-aided diagnose and human-machine-comrmmication systems.With the object localization (spine) in medical images as the application background,a statistical shape model based framework of Active Shape Model (ASM) is presented. After construction of the prior landmark models based on principal component analysis technique,structures of objects can be located by adjusting the model parameters in a learning way that it minimizes the distances of corresponding landmark points between model and object. A good overall match is obtained in a few iterations. Based on ASM,many localization experiments have been taken with actual X-ray medical images of human spines as the training sets. Those results show that ASM has a high level of veracity compared with the traditional methods of object location. In the meantime,ASM also has a high level of matching speed compared with the template-matching method. In order to improve the generality of this model,a new "two-dimensional sampling windows" method,in which the searching space is extended from the one- dimensional area to the two- dimensional area,is proposed in this paper. Besides ASM,a general approach towards object location using Active Appearance Model (AAM) is presented. AAM measures the current texture residuals and uses the model to predict changes to the current parameters,leading to a better fit. At last,the independent component analysis is used in the building procedure of Active Shape Model.